Robert Sachunsky

notebook crashes

2

I have not been able to isolate the cause, but the [example notebook](https://mybinder.org/v2/gh/poke1024/pyalign-demo/HEAD?filepath=example.ipynb) keeps crashing for me (with the kernel disconnecting after some steps). It looks somewhat racy/random, but I...

running from repo as working directory fails

1

If I `import pyalign` while still inside the source directory, I get the following: ``` RuntimeError: none of ['pyalign.algorithm.native', 'pyalign.algorithm.intel_avx2', 'pyalign.algorithm.apple_m1', 'pyalign.algorithm.generic'] is available ``` (The problem goes away when...

Although it would be possible to dive into the docstrings on the Python interpreter, I would very much like to view the generated API documentation (HTML output). Now, the Readme...

setup: add repo URL

2

The [project page on PyPI](https://pypi.org/project/pyalign/) lacks a backref to Github due to this. (Most images in the README do not render there, too.)

Support python3

3

Still works with Python 2 of course. (Tested with Python 2.7 and Python 3.6.7, OpenCV 4.2, SciPy 1.2.3, NumPy 1.16) Note: there are many more things that should be done...

The configuration for the `PrimaLayout` MRCNN Detectron2 model … https://github.com/Layout-Parser/layout-parser/blob/04e28168d820eea3a1ff1e098078323e7b48648b/src/layoutparser/models/detectron2/catalog.py#L57 … contains a setting `MODEL.ROI_HEADS.NUM_CLASSES: 7`. But LP's actual runtime config for the label map of the region types contains...

When I run with the weights and config linked in the Readme, Detectron2 cannot correctly resolve the base config file: ``` File "/lib/python3.7/site-packages/detectron2/config/config.py", line 46, in merge_from_file loaded_cfg = self.load_yaml_with_base(cfg_filename,...

serialization / templates: prefer re-use of segment identifiers, if existing (from XML parser)

1

When parsing ALTO or PAGE, you do already keep the identifiers of regions and lines. But the output throws this info away and generates vanilla block/line labels. It would be...

recognizer: Shapely problem, unspecific error message

4

When using the recognizer via CLI on PAGE files, I sometimes get failures like this: ``` Writing recognition results for GN 1771,4 (GN.A.158)/552_202cd_default.xml [01/24/24 12:28:34] ERROR Failed processing GN 1771,4...

recognition training: one_channel_mode not always promoted

12

In OCR-D, we [rely on the model metadata correctly specifying whether images should be binarized or not](https://github.com/OCR-D/ocrd_kraken/blob/802c6b0b76a3e75070c680aa3b19d36142decf4e/ocrd_kraken/recognize.py#L63) – via `one_channel_mode` property. However, there seems to be a gap in the...